Automated critical power limit estimation for natural convection-cooled research reactor core

https://doi.org/10.1016/j.anucene.2020.107332Get rights and content

Highlights

  • Design limit for critical heat flux ratio is evaluated.

  • Computational fluid dynamic simulation is carried out to yield power-flow relationship.

  • In-house companion program for batch control and pre- and post-processing of RELAP5 code is developed.

  • Critical power limit estimation method for a natural circulation-cooled research reactor core is presented.

  • Effect of critical heat flux correlation and core states on critical power is studied.

Abstract

The critical power limit for a natural circulation-cooled research reactor core is estimated by utilizing a system analysis code RELAP5/MOD3.3 and an in-house companion program RCPP. The variation range of thermal-hydraulic parameters is found from steady-state and transient simulation results. From literature, potential candidate correlations for critical power estimation were selected and the design limit critical heat flux ratio was evaluated for the chosen correlations based on error statistics. A computational fluid dynamic simulation was carried out to observe the power-equilibrium flow relationship and used to prepare the system code input. To ease a rigorous critical power estimation process, an in-house code was developed to automate batch run and output post-processing of system code. The critical power limit was found for each combination of major operation parameters. A critical power map was generated by merging each power limit value, and the effect of the correlation and core states were studied.

Introduction

Currently, more than two hundred research reactors are in operation worldwide, with varying core powers owing to their diverse utilization purposes*** (IAEA, 2019). The core heat is removed by forced cooling by pump or by natural convection. For a TRIGA (Training, Research, Isotopic Production General Atomic) reactor, it is reported that up to 2 MW of core thermal power can be safely removed by natural convection (IAEA, 2018). Based on this criterion, about 69% of the research reactors worldwide have cores that may be passively cooled. Some of the well-known research reactors with natural convective core cooling are TRIGA, SLOWPOKE, and MNSR. TRIGA was developed by General Atomics, and there are three design variations (Mark-I to Mark-III). For different utilization requirements, its design core power can be varied from several tens of kW to several MW. Most of the core utilizes rod-type U-ZrH fuel with top and bottom graphite reflectors (Stancar and Snoj, 2017). SLOWPOKE (Safe LOW-POwer Kritical Experiment) was designed by Atomic Energy of Canada Limited, where prototype (SLOWPOKE-1) and commercialized (SLOWPOKE-2) versions exists. The core thermal power of SLOWPOKE-2 is 20 kW, and rod-type fuel made of UO2 meat and Zircaloy-4 cladding is currently used (Townes and Hilborn, 1985). The MNSR (Miniature Neutron Source Reactor) was developed by China Institute of Atomic Energy. It has similar design concepts as SLOWPOKE-2. To remove 30 kW of core heat, the coolant flows into the core through the gap between the lower reflector and annular beryllium block. Its rod-type fuel utilizes U-Al alloy meat (Yongmao, 1986, Dawahra, 2014). Korea Atomic Energy Research Institute (KAERI) is also developing a low-power research reactor (LPRR) for training and educational purpose, where rod-type and low enriched uranium oxide fuel is utilized to generate 50 kW of thermal power (Kim, 2018).

Unlike forced cooled research reactors, the heat of the natural convection-cooled core is removed by buoyancy-induced upward flow, which means the flow rate is coupled with the power and not directly controllable. Since the core equilibrium flow rate is not linearly proportional (mass flow ∝ power1/3) to the thermal power, the amount of power increase cannot be fully compensated by the flow increase (Lewis, 1977). In the case of an operator malfunction or a reactivity-induced accident, the core power may increase. The power level then may reach the critical value (critical power) at which the local wall heat flux corresponds to the critical heat flux (CHF) and fuel failure can occur. The reactor should be tripped by any means before reaching this limit to bring the reactor to safety. This critical core power limit is used as a base value to establish safety limits by including conservatism and transient effects (IAEA, 2008, McGuire, 1995). Because finding this kind of limit is quite a rigorous task, this study developed an automated method to generate a power limit map for the natural convection-cooled research reactor core. First, the variation range of the major thermal-hydraulic parameters was found from steady-state and transient analysis results. A literature review was conducted to identify applicable CHF correlations and their design limits for estimating critical power. An in-house companion code RCPP (RELAP5 Companion Processor) was developed to control the batch run of the system code RELAP5/MOD3.3 and to post-process the results. The RELAP5/MOD3.3 is the system thermal hydraulic analysis code, which is widely used in the nuclear industry (USNRC, 2001). For each run, this code calculated thermal hydraulic variable values of the reactor core region necessary for the CHF ratio evaluation. Using the RELAP5/MOD3.3 and the RCPP, the critical power limit was found for each combination of the major operation parameters, namely, the core inlet temperature and pool height. Finally, a critical power map was created by merging the critical power limits for the range of operating conditions.

Section snippets

Problem description

In this study, a fuel assembly of the LPRR being developed by KAERI was modeled and analyzed to construct a critical core power limit map (Kim, 2018).

Correlation applicability

Table 3 summarizes candidate correlations after preliminary screening. The correlations were screened based on four selection criteria (general correlation, reputation, embedded in system analysis code, and has error statistics). An AECL lookup table (LUT) is a collection of CHF values for local thermal hydraulic conditions (pressure, mass flux, and quality) normalized for 8 mm tube geometry. Several correction factors were applied to consider geometric and flow-related effects (Groeneveld et

Computational fluid dynamic simulation

As seen in Fig. 1, the heat from the fuel rods is removed by coolant that flow through holes in the grid plate and core box wall. In this study, only the flow through grid plate holes were considered in the analysis, because reducing the number of possible inflow paths simplifies the problem and yields more conservative (higher) temperature distribution. Before preceding to the system analysis, a computational fluid dynamic (CFD) simulation was carried out using commercial code CFX 16.1 to

Critical core power limit map

The critical core power limit map was generated by combining predicted critical values for a wide range of thermal hydraulic boundary conditions (inlet temperature = 25 ~ 95 °C, pool height = 6 ~ 8 m) which covers the normal operation range. As shown in Table 7, maps were created for each combination of the CHF correlation and the core state. Fig. 19 shows contour maps of the resulting critical powers. Overall, as the pool water height is increased and the core inlet temperature are decreased,

Conclusions

In this study, a critical power limit map generation method was developed for the natural circulation-cooled research reactor core. In brief, this method is divided into two stages. The first stage deals with the constructing system code input for MCHFR calculation, which requires the correlation limit to be found from suitable CHF correlation and EHCF values. The next stage is about finding critical power values for given boundary conditions and convergence criteria. To automate this second

CRediT authorship contribution statement

Hyung Min Son: Conceptualization, Methodology, Software, Formal analysis, Writing - original draft. Jonghark Park: Validation, Formal analysis, Investigation, Visualization, Supervision, Project administration.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Acknowledgements

This work was conducted as a part of the Development of Research Reactor Technology project sponsored by the Ministry of Science and ICT of the Korean government.

References (32)

  • D.C. Groeneveld

    1986 AECL-UO critical heat flux lookup table

    Heat Transfer Eng.

    (1986)
  • IAEA, “Operational Limits and Conditions and Operating Procedures for Research Reactors,” Safety Guide, No. NS-G-4.4,...
  • IAEA, “TRIGA REACTOR MAIN SYSTEMS,” (https://ansn.iaea.org), retrieved at...
  • IAEA, Research Reactor Database (https://nucleus.iaea.org/RRDB), retrieved at...
  • Jordan Research and Training Reactor Final Safety Analysis Report, Rev.4, Jordan Atomic Energy Commission,...
  • M. Kalimullah

    Estimating hot channel factors for a generic MNSR using rodded fuel cooled by natural circulation

    Proceedings of 33rd International Reduced Enrichment for Research and Test Reactors

    (2011)
  • Cited by (0)

    View full text